Neurological assessment is a fundamental to the concept of comprehensive fetal monitoring. There are, however, no tests to reliably identify the fetus with neurological impairment. The current challenge is to increase the positive predictive value of testing for fetal neurological disease. Through our previously funded NIH/NINDS grant, we have progressed toward this goal by successfully developing and installing at the University of Arkansas for Medical Sciences (UAMS) the world's first biomagnetic-sensing system, SARA (SQUID Array for Reproductive Assessment), dedicated to fetal monitoring. Based on initial results, we believe fetal magnetoencephalography (fMEG) studies using the SARA system represent the correct approach to this challenge. By non-invasively detecting biomagnetic fields generated in the body, the 151 primary superconducting sensors provide the means for the investigation of maternalfetal parameters from conception to delivery. The measured spatial-temporal signals are a complex mixture from many different sources in the body. These signals must be efficiently separated into their constituent waveforms in order to be utilized for basic research and diagnostic purposes. After separation of sources, the signals can be analyzed and quantified. Also, the possible interactions of the signals must be determined. Prior to SARA, no device provided the means for complete investigation of maternal and fetal physiology. This system generates a high dimensional spatialtemporal dataset, the characteristics of which depend on the experimental task. The investigation of fetal brain development is mainly performed using certain sensory stimulation protocols such as evoked and steady state responses to sound and light.The optimal recording and analysis parameters of each task must be determined by comparing the efficiency of different stimulation parameters for fetal brain development. The success of these studies is dependent on the strength of the analysis tools. To program, validate, and optimize computational tools, efficient algorithms must be developed that can process large spatial-temporal datasets. [unreadable] [unreadable] This proposal addresses 2 maior tasks: 1) development of new analysis tools and technologies for fetal brain assessment, 2) implementation of a clinical study for assessment of neurologically normal and abnormal fetuses. [unreadable] [unreadable] The following objectives are directed towards these goals:1. improve data analysis techniques to attenuate maternal and fetal biomagnetio signals of interference, therefore enhancing extraction of reliable brain signals, and increase efficiency and speed of data analysis.2. Determine the validity of MEG signals by comparison of anatomical and physiological data3. Optimize the stimulation protocol for evoked fields to improve detection rate and determine possibleconfounding behavioral responses in the fetus 4. Determine the variability of response by repeated measurements on individual patients in a short period of time5. Determine the maturation of brain responses in-utero and compare the responses to newborns in a normal population to generate normative data6. Determine fetal brain development in abnormal fetuses